| 10 Aug 2026
Determination of δ13C-CO2 in ambient air by gas preconcentration-isotope ratio mass spectrometry: methodological development and optimization at sub-milliliter injection volumes
Abstract. The stable carbon isotope composition of atmospheric CO2 (δ13C-CO2) is a critical tracer for distinguishing the contributions of different carbon sources and sinks. However, the atmospheric background CO2 concentration is only approximately 420 ppm, which renders direct injection and measurement unachievable using conventional continuous-flow isotope ratio mass spectrometry and limits the widespread application of this technique in high-frequency monitoring scenarios. In this study, a gas preconcentration unit (PreCon) was coupled to an isotope ratio mass spectrometer (IRMS). Key parameters, including the trapping duration of the two-stage liquid nitrogen cold traps (T2/T3), sample injection volume, and sample vial pretreatment protocol, were systematically optimized, and methodological validation was conducted using multiple reference materials via multiple pretreatment pathways. The results demonstrated that the optimal trapping time for both T2 and T3 cold traps was 200 s, under which quantitative CO2 trapping was achieved without detectable isotopic fractionation. The system background signal accounted for approximately 0.4 % of the signal intensity of a typical sample, exerting no significant interference on the measurement results. Over the injection volume range of 0.5–5.0 mL, a strong linear correlation was observed between injection volume and signal response (R2 = 0.9998). The minimum effective injection volume was 0.5 mL (corresponding to approximately 8.9 nmol CO2), with a replicate measurement precision of 0.02 ‰ for δ13C. Helium flush was identified as the optimal pretreatment approach for air samples stored in Labco vials. Measurements via manual and automatic injection showed excellent consistency, and the results agreed well with the certified values of gas matrix reference materials. The maximum deviation among the three pretreatment pathways (PreCon, GasBench, and dual inlet (DI)) was 0.11 ‰, and the method enabled accurate determination of both solid carbonate and gas matrix reference materials. This proposed method achieves online high-precision δ13C measurement of atmospheric background CO2 at the milliliter scale, and provides reliable technical support for atmospheric carbon cycle tracing and isotopic monitoring of urban carbon emissions.
Overall assessment
This manuscript addresses a relevant analytical problem and presents a potentially useful two-stage cryogenic PreCon–CF-IRMS approach for δ13C-CO2 measurements from very small air volumes. The topic of the manuscript is well suited for publication in AMT however, I’d suggest significant revision that should substantially strengthen the analytical validation, clarify the calibration and traceability framework, improve the reproducibility of the experimental description, and better define the intended application and novelty of the method. The manuscript currently overstates the degree of methodological and metrological validation achieved. Several key claims are not sufficiently supported, including quantitative CO2 trapping, absence of isotopic fractionation and low measurement uncertainty. The reference-gas calibration and its traceability also need much clearer documentation. In addition, the manuscript does not sufficiently address the temporal dimension of atmospheric sampling (storage, stability and analytical throughput), despite invoking WMO/GAW and ICOS.
Major comments
1.Scientific context and framing of the analytical challenge
The scientific motivation in the Abstract and Introduction should be substantially revised. The manuscript currently conflates CO2 mole fraction, total sample amount, water interference and instrumental sensitivity. Atmospheric CO2 at ~400–420 ppm is not simply “far below the detection requirement” of CF-IRMS; the relevant limitation is rather the amount of CO2 available for analysis, together with matrix effects and the analytical configuration. Atmospheric CO2 can be measured by CF-IRMS from mL-scale air samples, and Yu and Lee (2020) specifically addressed atmospheric CO2 measurements. Leitner et al. (2023) likewise demonstrated atmospheric CO2 measurements using a GasBench II–cold trap–IRMS setup.
The presentation of DI-IRMS as poorly compatible with large-scale atmospheric monitoring should also be reconsidered. DI-IRMS remains an established high-precision approach for atmospheric δ13C-CO2 measurements. The authors should therefore reformulate the analytical problem and clarify precisely what limitation is addressed by the proposed PreCon T2/T3 approach and what its added value is relative to existing CF-IRMS and DI methods.
2. Experimental description is not sufficiently detailed for reproducibility
The Materials and Methods section provides the general setup but not enough information for faithful reproduction. Important details are missing for the PreCon system and T2/T3 configuration, gas flows and transfer sequence, syringes, Labco vials and stoppers, acid preparation, vacuum line, reaction manifold, quartz tubes, autosampler and software.
The authors should provide manufacturer/model information where relevant, specify the complete sampling and injection sequence, explain how gas is physically transferred between vial, PreCon and GasBench, and clarify whether samples are processed sequentially or in batches. Established procedures should be supported by appropriate methodological references.
3.Reference gases and calibration traceability need much more explicit documentation
The manuscript uses numerous gas and solid reference materials, but the information required to reconstruct the isotope calibration is insufficient. The authors should provide the assigned CO2 mole fraction and uncertainty, assigned δ13C-CO2 and uncertainty, source/certificate, preparation and certification procedure, and traceability to the relevant isotope scale.
The roles of primary/secondary reference materials and working gases should also be distinguished clearly. This is essential because the manuscript repeatedly claims traceability to VPDB.
4. Blank and vial-conditioning effects should be quantified more rigorously
The comparison between vacuum evacuation and He flushing is potentially useful, especially at sub-mL sample amounts. However, the impact of the blank should be evaluated isotopically, not only from its m/z 44 amplitude, since the blank has a different δ13C-CO2 composition from the samples.
The authors should quantify or estimate the resulting δ13C-CO2 bias and provide a more complete description of vial conditioning, including flushing conditions, delay before injection, storage time and possible effects of repeated sampling from the same vial.
5.Quantitative trapping and isotope fractionation require stronger evidence
The manuscript infers quantitative CO2 trapping from the linearity of the m/z 44 response with injection volume. However, a high R2 demonstrates linear response, not quantitative recovery or absence of breakthrough.
Similarly, the statement that no isotopic fractionation occurs during freeze–release cycles should be supported by explicit δ13C-CO2 values, replicate measurements and uncertainties. The choice of 200 s for T2 and T3 should also be quantitatively justified, including whether shorter trapping times could provide adequate recovery and isotopic performance.
6. Calibration and validation of trueness are not independent
The calibration appears to use the same reference materials to establish the calibration relationship and to evaluate its performance. In this context, R2=1 cannot be considered evidence of trueness.
The authors should distinguish calibration materials from independent validation materials, or provide another independent validation strategy. Slope, intercept, residuals and bias should be reported in addition to R2, and the uncertainties of the reference materials should be incorporated into the evaluation of bias and measurement uncertainty.
7.The role and added value of the carbonate experiments should be clarified
Carbonate digestion with phosphoric acid and subsequent CF-IRMS/DI analysis are well-established procedures, including for very small samples (e.g. Fiebig et al., 2005). The authors should therefore clarify what is specifically demonstrated by the carbonate experiments beyond the established carbonate-to- CO2 conversion.
The comparison among PreCon, GasBench and DI is also complicated because the different pathways use different sample masses, vessels and preparation procedures. The authors should better discuss this limitation and explain what aspect of the analytical method is actually being validated.
8. Analytical terminology and statistical treatment need revision
Precision, repeatability, stability, accuracy, trueness, bias and measurement uncertainty are used a lot. These terms should be defined.
Several conclusions also imply statistical significance (“no significant difference”, “no detectable fractionation”, “excellent consistency”) without providing statistical tests or uncertainty estimates. In addition, the reported 0.02 ‰ precision does not appear, from Table 3, to be a conventional repeatability estimate at a fixed injection volume. The calculation and terminology should be reconsidered throughout.
9. Atmospheric relevance, storage and analytical throughput need clarification
The manuscript invokes WMO/GAW and ICOS and discusses high-frequency atmospheric monitoring, but the temporal dimension of the measurement workflow is largely absent. It is unclear whether the method is intended for truly online sequential analysis or for discrete samples collected and analysed later.
For stored samples, stability of CO2 amount and δ13C-CO2 over time is an important part of the analytical method; for online measurements, the complete cycle time and effective throughput should be reported.
10. Discussion and literature context are too limited
The Results and Discussion section contains relatively few references to previous methodological work. As a result, much of the discussion remains descriptive (“observation → selected condition”) rather than analytical or comparative.
The authors should relate their observations to previous work on cryogenic trapping, vial conditioning, atmospheric CO2 analysis, blanks, isotope fractionation, carbonate digestion and calibration. The recent literature should be discussed more systematically.
Specific comments